Development of a machine learning-based acuity score prediction model for virtual care settings

Machine learning outperforms the Canadian Triage and Acuity Scale (CTAS) in predicting need for early critical care

STUDY OBJECTIVE

This study investigates the potential to improve emergency department (ED) triage using machine learning models by comparing their predictive performance with the Canadian Triage Acuity Scale (CTAS) in identifying the need for critical care within 12 h of ED arrival.

METHODS

Three machine learning models (LASSO regression, gradient-boosted trees, and a deep learning model with embeddings) were developed using retrospective data from 670,841 ED visits to the Jewish General Hospital from June 2012 to Jan 2021. The model outcome was the need for critical care within the first 12 h of ED arrival. Metrics, including the areas under the receiver-operator characteristic curve (ROC) and precision-recall curve (PRC) were used for performance evaluation. Shapley additive explanation scores were used to compare predictor importance.

RESULTS

The three machine learning models (deep learning, gradient-boosted trees and LASSO regression) had areas under the ROC of 0.926 ± 0.003, 0.912 ± 0.003 and 0.892 ± 0.004 respectively, and areas under the PRC of 0.27 ± 0.01, 0.24 ± 0.01 and 0.23 ± 0.01 respectively. In comparison, the CTAS score had an area under the ROC of 0.804 ± 0.006 and under the PRC of 0.11 ± 0.01. The predictors of most importance were similar between the models.

CONCLUSIONS

Machine learning models outperformed CTAS in identifying, at the point of ED triage, patients likely to need early critical care. If validated in future studies, machine learning models such as the ones developed here may be considered for incorporation in future revisions of the CTAS triage algorithm, potentially improving discrimination and reliability.

Improving triage performance in emergency departments using machine learning and natural language processing: a systematic review

BACKGROUND

In Emergency Departments (EDs), triage is crucial for determining patient severity and prioritizing care, typically using the Manchester Triage Scale (MTS). Traditional triage systems, reliant on human judgment, are prone to under-triage and over-triage, resulting in variability, bias, and incorrect patient classification. Studies suggest that Machine Learning (ML) and Natural Language Processing (NLP) could enhance triage accuracy and consistency. This review analyzes studies on ML and/or NLP algorithms for ED patient triage.

METHODS

Following Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines, we conducted a systematic review across five databases: Web of Science, PubMed, Scopus, IEEE Xplore, and ACM Digital Library, from their inception of each database to October 2023. The risk of bias was assessed using the Prediction model Risk of Bias Assessment Tool (PROBAST). Only articles employing at least one ML and/or NLP method for patient triage classification were included.

RESULTS

Sixty studies covering 57 ML algorithms were included. Logistic Regression (LR) was the most used model, while eXtreme Gradient Boosting (XGBoost), decision tree-based algorithms with Gradient Boosting (GB), and Deep Neural Networks (DNNs) showed superior performance. Frequent predictive variables included demographics and vital signs, with oxygen saturation, chief complaints, systolic blood pressure, age, and mode of arrival being the most retained. The ML algorithms showed significant bias risk due to critical bias assessment in classification models.

CONCLUSION

NLP methods improved ML algorithms' classification capability using triage nursing and medical notes and structured clinical data compared to algorithms using only structured data. Feature engineering (FE) and class imbalance correction methods enhanced ML workflows' performance, but FE and eXplainable Artificial Intelligence (XAI) were underexplored in this field. Registration and funding. This systematic review has been registered (registration number: CRD42024604529) in the International Prospective Register of Systematic Reviews (PROSPERO) and can be accessed online at the following URL: https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=604529 . Funding for this work was provided by the National Council for Scientific and Technological Development (CNPq), Brazil.

Evaluating the Reliability of a Remote Acuity Prediction Tool in a Canadian Academic Emergency Department

STUDY OBJECTIVE

There is increasing interest in harnessing artificial intelligence to virtually triage patients seeking care. The objective was to examine the reliability of a virtual machine learning algorithm to remotely predict acuity scores for patients seeking emergency department (ED) care by applying the algorithm to retrospective ED data.

METHODS

This was a retrospective review of adult patients conducted at an academic tertiary care ED (annual census 65,000) from January 2021 to August 2022. Data including ED visit date and time, patient age, sex, reason for visit, presenting complaint and patient-reported pain score were used by the machine learning algorithm to predict acuity scores. The algorithm was designed to up-triage high-risk complaints to promote safety for remote use. The predicted scores were then compared to nurse-led triage scores previously derived in real time using the electronic Canadian Triage and Acuity Scale (eCTAS), an electronic triage decision-support tool used in the ED. Interrater reliability was estimated using kappa statistics with 95% confidence intervals (CIs).

RESULTS

In total, 21,469 unique ED patient encounters were included. Exact modal agreement was achieved for 10,396 (48.4%) patient encounters. Interrater reliability ranged from poor to fair, as estimated using unweighted kappa (0.18, 95% CI 0.17 to 0.19), linear-weighted kappa (0.25, 95% CI 0.24 to 0.26), and quadratic-weighted kappa (0.36, 95% CI 0.35 to 0.37) statistics. Using the nurse-led eCTAS score as the reference, the machine learning algorithm overtriaged 9,897 (46.1%) and undertriaged 1,176 (5.5%) cases. Some of the presenting complaints under-triaged were conditions generally requiring further probing to delineate their nature, including abnormal lab/imaging results, visual disturbance, and fever.

CONCLUSION

This machine learning algorithm needs further refinement before being safely implemented for patient use.